1. The Field of the Invention
The present invention relates to event notification technology. More specifically, the present invention relates to methods, systems and computer program products for performing event notification based on subscriptions to a notification channel.
2. Background and Related Art
Most would agree that intelligent decision-making requires access to relevant and timely information. Computers have greatly enhanced our ability to access such information, resulting in this modern era often being termed “the information age”.
Some access to information is synchronous or “pull-oriented”. In other words, information is accessed each time a request for the information is submitted. There are some cases in which it may be inefficient to synchronously access information. For example, some information may be unexpected, yet still relevant. Of course, a user or application would not typically know to request information that was unexpected. Accordingly, technologies have developed that notify users or applications when an event of interest occurs asynchronously or, in other words, without requiring a corresponding request to inquire as to whether the event occurred.
Events of interest might include, for example, that a computer error has occurred since one might want to save data if one knows that the computer is becoming unstable. Also, one might want to know if a particular data field has a particular value or if a combination of data fields has a particular combination of data values. For example, an investor might want to know if the price of a stock drops below a particular value during a particular time of the year. Also, one might want to know of other asynchronous events such as, for example, the receipt of an e-mail from a particular sender.
One particular application that may be notified of the occurrence of events is an event log. Conventional event logs typically log a large number of events regarding a wide variety of subjects, regardless of whether any application or user is interested in any given event. If the user is interested in determining a particular sequence of events, then the user would typically view the entire log and visually track the events of interest to draw appropriate conclusions. For example, suppose that a computer systems crashes. A network administrator may view the event log, sift through all of the information to distinguish the relevant from the irrelevant, and then draw a conclusion based on the relevant information as to why the computer system crashed.
It is well recognized that the human brain is capable of performing astounding mental processes. For example, human users are often able to easily distinguish the relevant information from the irrelevant, and form judgments based on the relevant, even if there is very little consistent structure in how the information is represented. Computer systems are, however, more limited in their ability to interpret information that does not have a predictable structure.
Accordingly, log entries have some predictable structure that allows computers some ability to interpret information. For example, one conventional event log technology assigns a limited number of fields for each event entry. Such fields may include, for example, an event ID, an event source ID, an event type (e.g., information, warning, or error in increasing severity levels), and a description. The number and types of fields used to represent an event entry are not typically extensible. Accordingly, the type of event information that may be recorded is quite restricted.
One conventional method that somewhat reduces this restriction is to have applications draft a description that has some predetermined structure in the description field of the event entry. Then, applications that understand that structure may sift through the description and extract the intended meaning using the predetermined structure. This requires, however, that the application that interprets the description have an understanding of the predetermined structure. Conventionally, each application uses its own predetermined structure when writing the event description. Accordingly, the event description for every event may still be difficult to interpret due to that lack of standardization regarding how the description field should be structured.
Also, since the event log records all events, regardless of whether any g applications or users are interested in the event, precious computer processing and memory A resource may be wasted writing an event entry that is of no interest.
Accordingly, what is desired are event notification technologies that allow broad-based interpretation of events, and that reduces processing and memory resources when processing event of lesser interest.